JP2012125297A - Ultrasonic diagnostic apparatus, image processor, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform accurate positioning between volume data.SOLUTION: In an ultrasonic diagnostic apparatus, an object cross section-receiving unit 17a receives, in two volume data items, or ultrasonic volume data items in which at least one of them is generated based on reflection waves of ultrasonic waves, two cross sections (object cross sections) including a common structure from an operator via an input device 3. A calculation unit 17b divides, into a plurality of small fragments, a first cross section that is either one cross section of the two object cross sections received by the object cross section-receiving unit 17a, and calculates the amounts of features of a plurality of small fragments. An extraction unit 17c extracts ROI for positioning the two volume data items from the plurality of small fragments, based on the respective amount of features of the plurality of small fragments calculated by the calculation unit 17b. A positioning unit 17e performs the positioning of the two volume data items using the ROI extracted by the extraction unit 17c.

Description

  Embodiments described herein relate generally to an ultrasonic diagnostic apparatus, an image processing apparatus, and a program.

  Conventionally, a three-dimensional ultrasonic image (volume data) in real time using a mechanical scan probe that swings a plurality of transducers arranged in a row or a two-dimensional ultrasound probe in which a plurality of transducers are arranged in a grid pattern. An ultrasonic diagnostic apparatus capable of generating and displaying a signal has been put into practical use.

  In addition, in order to compare and interpret volume data generated at different times, development of a three-dimensional alignment technique for aligning a plurality of volume data is in progress. Here, if the deviation between the two volume data to be compared can be simply corrected by translational movement or rotational movement, the entire area of these two volume data is compared and aligned. Is easy. However, due to factors such as the volume data collection method, the patient's posture during imaging, and the patient's treatment, there are changes in luminance distribution, shape changes, shape distortions, etc. There are often differences. Therefore, it is difficult and inefficient to perform alignment by comparing the entire areas of the two volume data.

  For this reason, in the conventional three-dimensional alignment technology, a region of interest (ROI) for alignment is set in each volume data, and alignment of the two volume data is performed using the set ROI. It is done.

  For example, an ultrasonic diagnostic apparatus that performs three-dimensional alignment of two volume data by setting an anatomically characteristic small region as an ROI is known. Such an ultrasonic diagnostic apparatus sets a small region manually extracted by a user or a small region automatically extracted by calculating a feature amount such as an average information amount (entropy) as an ROI. By performing the three-dimensional alignment process using the ROI set by such a process, the doctor can perform comparative interpretation of at least two volume data in which the interpretation target part is aligned.

JP 2009-112468 A

  However, the ROI set by the conventional technique described above may be difficult to correctly align depending on the set position.

  The ultrasonic diagnostic apparatus according to the embodiment includes a reception unit, a calculation unit, an extraction unit, and an alignment unit. The receiving unit receives two target cross sections including a common structure from an operator via a predetermined input unit in two volume data, at least one of which is volume data generated based on an ultrasonic reflected wave. . The calculation unit divides the first cross-section that is one of the two target cross-sections received by the receiving unit or the first region including the first cross-section into a plurality of small fractions, The feature amount of each of the small fractions is calculated. The extraction unit extracts a region of interest for aligning the two volume data from the plurality of small fractions based on the feature amounts of the plurality of small fractions calculated by the calculation unit. The alignment unit aligns the two volume data using the region of interest extracted by the extraction unit.

FIG. 1 is a diagram for explaining the configuration of the ultrasonic diagnostic apparatus according to the present embodiment. FIG. 2 is a diagram for explaining the image generation unit according to the present embodiment. FIG. 3A is a diagram (1) for explaining the importance of ROI setting in alignment. FIG. 3B is a diagram (2) for explaining the importance of the ROI setting in the alignment. FIG. 4 is a diagram (1) for explaining the volume data processing unit shown in FIG. FIG. 5 is a diagram (2) for explaining the volume data processing unit shown in FIG. FIG. 6 is a diagram for explaining an example of the processing result of the alignment unit. FIG. 7 is a flowchart for explaining the ROI setting process of the ultrasonic diagnostic apparatus according to this embodiment. FIG. 8 is a flowchart for explaining the alignment process of the ultrasonic diagnostic apparatus according to the present embodiment. FIG. 9 is a diagram for explaining a modification according to the present embodiment.

  Hereinafter, embodiments of an ultrasonic diagnostic apparatus will be described in detail with reference to the accompanying drawings.

(Embodiment)
First, the configuration of the ultrasonic diagnostic apparatus according to the present embodiment will be described. FIG. 1 is a diagram for explaining the configuration of the ultrasonic diagnostic apparatus according to the present embodiment. As shown in FIG. 1, the ultrasonic diagnostic apparatus according to the first embodiment includes an ultrasonic probe 1, a monitor 2, an input device 3, and an apparatus main body 10. The apparatus main body 10 is connected to the external apparatus 4 via the network 100.

  The ultrasonic probe 1 includes a plurality of piezoelectric vibrators, and the plurality of piezoelectric vibrators generate ultrasonic waves based on a drive signal supplied from a transmission / reception unit 11 included in the apparatus main body 10 described later. The ultrasonic probe 1 receives a reflected wave from the subject P and converts it into an electrical signal. The ultrasonic probe 1 includes a matching layer provided on the piezoelectric vibrator, a backing material that prevents propagation of ultrasonic waves from the piezoelectric vibrator to the rear, and the like. The ultrasonic probe 1 is detachably connected to the apparatus main body 10.

  When ultrasonic waves are transmitted from the ultrasonic probe 1 to the subject P, the transmitted ultrasonic waves are reflected one after another at the discontinuous surface of the acoustic impedance in the body tissue of the subject P, and the ultrasonic probe is used as a reflected wave signal. 1 is received by a plurality of piezoelectric vibrators. The amplitude of the received reflected wave signal depends on the difference in acoustic impedance at the discontinuous surface where the ultrasonic wave is reflected. Note that the reflected wave signal when the transmitted ultrasonic pulse is reflected on the moving blood flow or the surface of the heart wall depends on the velocity component of the moving body in the ultrasonic transmission direction due to the Doppler effect. And undergoes a frequency shift.

  Here, the ultrasound probe 1 according to the present embodiment is an ultrasound probe capable of scanning the subject P in two dimensions with ultrasound and scanning the subject P in three dimensions. Specifically, the ultrasonic probe 1 according to the present embodiment swings a plurality of ultrasonic transducers that scan the subject P in two dimensions at a predetermined angle (swing angle), so that the subject P Is a mechanical scan probe that scans in three dimensions.

  In this embodiment, the ultrasonic probe 1 is a two-dimensional ultrasonic probe that can ultrasonically scan the subject P in three dimensions by arranging a plurality of ultrasonic transducers in a matrix. Even in some cases, it is applicable. The two-dimensional ultrasonic probe can scan the subject P in two dimensions by focusing and transmitting ultrasonic waves.

  The input device 3 is connected to the device main body 10 via an interface unit 19 described later. The input device 3 includes a mouse, a keyboard, a button, a panel switch, a touch command screen, a foot switch, a trackball, and the like, accepts various setting requests from an operator of the ultrasonic diagnostic apparatus, and accepts them to the apparatus body 10. Transfer various setting requests.

  For example, the input device 3 according to the present embodiment receives from the operator an instruction regarding designation of two volume data for performing comparative interpretation and an alignment process between the two designated volume data. The details of various instructions input by the operator using the input device 3 when performing alignment and comparative interpretation are described later.

  The monitor 2 displays a GUI (Graphical User Interface) for an operator of the ultrasonic diagnostic apparatus to input various setting requests using the input device 3, or displays an ultrasonic image generated in the apparatus main body 10. To do.

  For example, the monitor 2 according to the present embodiment displays a GUI referred to by the operator in order to perform alignment between two volume data. Note that the GUI referred to by the operator when performing alignment will be described in detail later.

  The external device 4 is a device connected to the device main body 10 via an interface unit 19 described later. For example, the external device 4 is a PACS (Picture Archiving and Communication System) database, which is a system for managing various medical image data, or an electronic medical record system database for managing an electronic medical record with attached medical images. . Alternatively, the external device 4 is various medical image diagnostic apparatuses other than the ultrasonic diagnostic apparatus according to the present embodiment, such as an X-ray CT (Computed Tomography) apparatus and an MRI (Magnetic Resonance Imaging) apparatus.

  That is, the apparatus main body 10 according to the present embodiment can acquire various medical images unified in an image format conforming to DICOM (Digital Imaging and Communications in Medicine) from the external apparatus 4 via the interface unit 19. . Specifically, the apparatus main body 10 according to the present embodiment sends the volume data to be compared with the volume data (three-dimensional ultrasonic image) generated by the own apparatus via the interface unit 19 to the interface unit 19. Via the external device 4.

  The apparatus main body 10 is an apparatus that generates an ultrasonic image based on the reflected wave received by the ultrasonic probe 1. Specifically, the apparatus main body 10 according to the present embodiment is an apparatus that can generate a three-dimensional ultrasonic image (volume data) based on three-dimensional reflected wave data received by the ultrasonic probe 1. As shown in FIG. 1, the apparatus body 10 includes a transmission / reception unit 11, a B-mode processing unit 12, a Doppler processing unit 13, an image generation unit 14, an image memory 15, an internal storage unit 16, and volume data processing. A unit 17, a control unit 18, and an interface unit 19.

  The transmission / reception unit 11 includes a trigger generation circuit, a delay circuit, a pulsar circuit, and the like, and supplies a drive signal to the ultrasonic probe 1. The pulsar circuit repeatedly generates rate pulses for forming transmission ultrasonic waves at a predetermined rate frequency. The delay circuit also sets the delay time for each piezoelectric vibrator necessary for determining the transmission directivity by focusing the ultrasonic wave generated from the ultrasonic probe 1 into a beam shape, and for each rate pulse generated by the pulser circuit. Give to. The trigger generation circuit applies a drive signal (drive pulse) to the ultrasonic probe 1 at a timing based on the rate pulse. In other words, the delay circuit arbitrarily adjusts the transmission direction from the piezoelectric vibrator surface by changing the delay time given to each rate pulse.

  The transmission / reception unit 11 has a function capable of instantaneously changing a transmission frequency, a transmission drive voltage, and the like in order to execute a predetermined scan sequence based on an instruction from the control unit 18 described later. In particular, the change of the transmission drive voltage is realized by a linear amplifier type transmission circuit capable of instantaneously switching its value or a mechanism for electrically switching a plurality of power supply units.

  The transmission / reception unit 11 includes an amplifier circuit, an A / D converter, an adder, and the like, and performs various processes on the reflected wave signal received by the ultrasonic probe 1 to generate reflected wave data. The amplifier circuit amplifies the reflected wave signal for each channel and performs gain correction processing. The A / D converter performs A / D conversion on the gain-corrected reflected wave signal and gives a delay time necessary for determining reception directivity to the digital data. The adder performs an addition process of the reflected wave signal processed by the A / D converter to generate reflected wave data. By the addition processing of the adder, the reflection component from the direction corresponding to the reception directivity of the reflected wave signal is emphasized.

  As described above, the transmission / reception unit 11 controls transmission directivity and reception directivity in ultrasonic transmission / reception. Here, the transmission / reception unit 11 according to the present embodiment transmits a three-dimensional ultrasonic beam from the ultrasonic probe 1 to the subject P, and three-dimensionally receives the three-dimensional reflected wave signal received by the ultrasonic probe 1. The reflected wave data is generated.

  The B-mode processing unit 12 receives the reflected wave data from the transmission / reception unit 11 and performs logarithmic amplification, envelope detection processing, and the like to generate data (B-mode data) in which the signal intensity is expressed by brightness. .

  Here, the B-mode processing unit 12 can change the frequency band to be visualized by changing the detection frequency. Further, the B-mode processing unit 12 can perform detection processing with two detection frequencies in parallel on one reflected wave data.

  By using the function of the B-mode processing unit 12, the contrast agent (microbubbles, bubbles) is reflected from the reflected wave data of the subject P into which the contrast agent is injected in contrast harmonic imaging (CHI). The reflected wave data (harmonic data or frequency division data) used as the source and the reflected wave data (fundamental wave data) using the tissue in the subject P as the reflection source can be separated. In other words, the B-mode processing unit 12 can generate B-mode data for generating a contrast image together with B-mode data for generating a tissue image.

  Further, by using the function of the B-mode processing unit 12, in the tissue harmonic imaging (THI), the harmonic data or the divided frequency data is separated from the reflected wave data of the subject P. B-mode data for generating a tissue image from which components are removed can be generated.

  The Doppler processing unit 13 performs frequency analysis on velocity information from the reflected wave data received from the transmission / reception unit 11, extracts blood flow, tissue, and contrast agent echo components due to the Doppler effect, and mobile body information such as average velocity, dispersion, and power. Is generated for multiple points (Doppler data).

  Note that the B-mode processing unit 12 and the Doppler processing unit 13 according to the present embodiment can process both two-dimensional reflected wave data and three-dimensional reflected wave data. That is, the B-mode processing unit 12 according to the present embodiment can generate three-dimensional B-mode data from the three-dimensional reflected wave data. Specifically, the B-mode processing unit 12 according to the present embodiment can generate three-dimensional B-mode data during normal B-mode imaging, contrast harmonic imaging, and tissue harmonic imaging. In addition, the Doppler processing unit 13 according to the present embodiment can generate three-dimensional Doppler data from the three-dimensional reflected wave data.

  The image generation unit 14 generates an ultrasound image from the data generated by the B mode processing unit 12 and the Doppler processing unit 13. That is, the image generation unit 14 generates a B-mode image in which the intensity of the reflected wave is represented by luminance from the B-mode data generated by the B-mode processing unit 12. Specifically, the image generation unit 14 generates a three-dimensional B board image from the three-dimensional B mode data generated by the B mode processing unit 12.

  In addition, the image generation unit 14 generates a color Doppler image as an average velocity image, a dispersed image, a power image, or a combination image representing moving body information from the Doppler data generated by the Doppler processing unit 13. Specifically, the image generation unit 14 generates a three-dimensional color Doppler image from the three-dimensional Doppler data generated by the Doppler processing unit 13.

  Hereinafter, the three-dimensional B-mode image and the three-dimensional color Doppler image generated by the image generation unit 14 are collectively referred to as “ultrasonic volume data”.

  The image generation unit 14 can generate various images for displaying the generated ultrasonic volume data on the monitor 2. Specifically, the image generation unit 14 can generate an MPR (Multi Planar Reconstructions) image or a rendering image from the ultrasonic volume data. FIG. 2 is a diagram for explaining the image generation unit according to the present embodiment.

  That is, as shown in FIG. 2, the ultrasound probe 1 performs a three-dimensional scan of the ultrasound on the imaging region of the subject P, so that the transmission / reception unit 11 generates ultrasound volume data. Then, the image generation unit 14 displays, as an image for displaying the ultrasonic volume data on the monitor 2, for example, in accordance with an instruction from the operator, as shown in FIG. A rendering image when a contact surface with respect to one subject P is a viewpoint or a rendering image when an arbitrary place is a viewpoint are generated. Note that the image generation unit 14 according to the present embodiment can perform MPR image and rendering image generation processing for volume data generated by devices other than its own device.

  The image generation unit 14 can also generate a composite image in which character information, scales, body marks, and the like of various parameters are combined with the ultrasonic image.

  Returning to FIG. 1, the image memory 15 is a memory for storing an ultrasonic image generated by the image generation unit 14. The image memory 15 can also store data generated by the B-mode processing unit 12 and the Doppler processing unit 13.

  The internal storage unit 16 stores a control program for performing ultrasonic transmission / reception, image processing and display processing, diagnostic information (for example, patient ID, doctor's findings, etc.), various data such as a diagnostic protocol and various body marks. To do. The internal storage unit 16 is also used for storing images stored in the image memory 15 as necessary.

  Further, the internal storage unit 16 is also used for storing various medical images transferred from the external device 4. Specifically, the internal storage unit 16 stores volume data transferred from the external device 4. For example, the internal storage unit 16 stores a three-dimensional X-ray CT image (hereinafter referred to as X-ray CT volume data), a three-dimensional MRI image (hereinafter referred to as MRI volume data), and other ultrasonic diagnosis. The ultrasonic volume data generated by the apparatus is stored. The data stored in the internal storage unit 16 can be transferred to an external peripheral device (external device 4) via an interface unit 19 described later.

  In the present embodiment, image data (volume data) desired by the operator is stored in the internal storage unit 16 via a storage medium such as a flexible disk (FD), CD-ROM, MO, or DVD. Even if it exists, it is applicable. Further, the present embodiment is applicable even when a storage device that stores image data (volume data) desired by the operator is installed other than the internal storage unit 16.

  The volume data processing unit 17 aligns two ultrasonic volume data stored in the image memory 15 and aligns the ultrasonic volume data stored in the image memory 15 and various volume data stored in the internal storage unit 16. Perform processing. Here, as shown in FIG. 1, the volume data processing unit 17 performs an alignment process, as shown in FIG. 1, a target cross-section receiving unit 17a, a calculation unit 17b, an extraction unit 17c, an installation unit 17d, and an alignment unit. 17e. The processing performed by the volume data processing unit 17 will be described in detail later.

  The control unit 18 controls the entire processing of the ultrasonic diagnostic apparatus. Specifically, the control unit 18 is based on various setting requests input from the operator via the input device 3 and various control programs and various data read from the internal storage unit 16. Controls the processing of the processing unit 12, Doppler processing unit 13, image generation unit 14, and volume data processing unit 17. The control unit 18 also includes an ultrasonic image stored in the image memory 15, various images stored in the internal storage unit 16, a GUI for performing processing by the volume data processing unit 17, and processing performed by the volume data processing unit 17. Control is performed so that the result is displayed on the monitor 2. Further, the control unit 18 performs control so that volume data received from the operator via the input device 3 is transferred from the external device 4 to the internal storage unit 16 via the network 100 and the interface unit 19.

  The interface unit 19 is an interface to the input device 3, the network 100, and the external device 4. Various setting information and various instructions from the operator received by the input device 3 are transferred to the control unit 18 by the interface unit 19. Also, the transfer request for image data received by the input device 3 from the operator is notified to the external device 4 via the network 100 by the interface unit 19. The image data transferred by the external device 4 is stored in the internal storage unit 16 by the interface unit 19.

  The overall configuration of the ultrasonic diagnostic apparatus according to this embodiment has been described above. Under such a configuration, the ultrasonic diagnostic apparatus according to the present embodiment generates ultrasonic volume data, and performs alignment between the generated ultrasonic volume data and other volume data.

  Here, when two volume data generated at different times are comparatively interpreted, if the deviation between the volume data can be corrected simply by translational movement or rotational movement, all of these two volume data are It is easy to perform alignment by comparing regions.

  However, it is difficult and inefficient to perform alignment by comparing the entire areas of the two volume data for the following reasons. For example, when a surgical procedure such as radiofrequency ablation (RFA) or excision is performed, the structure of the observation target region depicted in the two ultrasound volume data to be compared is It changes with. Further, for example, when the ultrasonic image capturing modes are different, such as B-mode imaging, tissue harmonic imaging, and contrast imaging, the luminance distributions of the two ultrasonic volume data to be compared are different.

  In clinical practice, ultrasonic volume data is often compared with volume data generated by other medical image diagnostic apparatuses such as an X-ray CT apparatus and an MRI apparatus. For example, when performing RFA, X-ray CT volume data used to determine the ablation site by RFA is compared with ultrasonic volume data generated to confirm the ablation site of RFA immediately before treatment. There is a need to. Even in such a case, the luminance distributions of the two volume data to be compared are different.

  In addition, depending on the breathing of the subject P and the posture at the time of imaging, there is a case where an organ to be observed is deformed or a relative position between the organ to be observed and surrounding organs is changed. In particular, in the imaging of ultrasonic images, the degree of freedom of the posture of the subject P is higher than in the imaging of X-ray CT images and MRI images, and the deformation of the organ to be observed, the organ to be observed, and Often changes in the relative position with surrounding organs occur. Further, in the imaging of an ultrasonic image, depending on the installation position of the ultrasonic probe 1 and the degree of compression by the ultrasonic probe 1, the deformation of the organ to be observed, or the relative relationship between the organ to be observed and the surrounding organs. A change in position may occur.

  In this way, when two volume data are aligned, there are many differences between the volume data, such as a change in luminance distribution, a change in shape, and a distortion in shape. For this reason, it is difficult and inefficient to align the volume data over the entire area of each volume data. Here, for a doctor who performs comparative interpretation, at least the site to be observed only needs to be accurately aligned. Therefore, when aligning two volume data, a region of interest (ROI) for alignment is set in each volume data, and the two ROIs are set using the set ROI. Alignment is performed.

  However, depending on how the ROI is placed, correct alignment may be difficult. 3A and 3B are diagrams for explaining the importance of ROI installation in alignment.

  The graphs shown in FIGS. 3A and 3B plot the mutual information amount between the ROIs of each volume data calculated by rotating one of the two volume data to be compared three-dimensionally by a predetermined angle. It is a thing. The mutual information amount will be described later in detail.

  Here, FIGS. 3-1 (A) and (B) show graphs when one ROI is installed in two volume data to be compared. The example shown in (A) of FIG. 3A shows a graph in which there is only one rotational maximum between the two volume data to be compared, and therefore there is only one maximum point where the alignment is optimal. ing. That is, in the example shown in FIG. 3A, the shape of the graph has a mountain shape with one vertex, and it is easy to reach a correct result when the alignment process is performed. On the other hand, in the example shown in FIG. 3B, since there is a difference in shape and the like in addition to the rotational deviation between the two volume data to be compared, the pseudo-maximal at which the alignment is apparently optimal. The graph when there are a plurality of points is shown. That is, in the example shown in FIG. 3B, the shape of the graph is a mountain shape having a plurality of vertices, and depending on how the ROI is placed, the alignment processing does not reach a correct result. ing.

  FIG. 3-2 shows the mutual information calculated by sequentially increasing the ROI from 1 to 4 according to the operator's designation. That is, (A) in FIG. 3-2 is a graph when one ROI is installed, and (B) in FIG. 3-2 is a graph when two ROIs are installed. (C) is a graph when three ROIs are installed, and (D) of FIG. 3-2 is a graph when four ROIs are installed. In the example illustrated in FIG. 3B, the number of pseudo maximum points is reduced by increasing the ROI.

  However, the positioning process between volume data using ROI is to obtain the position (optimum solution) where the mutual information amount of each ROI is maximized by the optimization process. In the example shown in FIG. Since the ROI is installed along the direction toward the true maximum point, the number of pseudo maximum points is reduced to show a case where true alignment is performed.

  For example, in the graph shown in FIG. 3D, when the ROI is installed at a flat position, the optimal solution cannot be reached, and only the local solution may be reached.

  Therefore, the volume data processing unit 17 shown in FIG. 1 performs the alignment process shown below. First, the operator designates two volume data for comparative interpretation.

  Here, a specific example of a use case in which the alignment process is performed will be described. For example, when liver cancer is treated, an X-ray CT image (X-ray CT volume data) using a contrast agent is taken as a preoperative examination for determining a treatment policy. For example, X-ray CT volume data of the liver is generated by CT (CTA) under hepatic arteriography or CT (CTAP) under transarterial portal angiography. Accordingly, the doctor confirms the position and size of the liver cancer and determines a surgical procedure such as surgical resection, RFA, or TAE (hepatic artery embolization therapy). For example, a doctor decides to ablate liver cancer by RFA.

  When performing RFA, a doctor punctures a puncture needle to a lesion site while confirming a target lesion by an ultrasonic image in real time, and then radiates radio waves from the puncture needle. Therefore, the doctor performs imaging of liver cancer determined to perform RFA using the X-ray CT volume data using an ultrasonic diagnostic apparatus. For example, the ultrasonic diagnostic apparatus generates ultrasonic volume data by CHI or THI. Accordingly, the doctor determines the approach position and the ablation area of the ultrasonic probe 1 having the attachment of the puncture needle that radiates radio waves. When performing such preliminary examination, it is necessary to align the X-ray CT volume data at the time of determining the technique with the ultrasonic volume data at the time of preliminary examination (use case 1).

  An inspection using an ultrasonic diagnostic apparatus is performed immediately before RFA. That is, the doctor generates ultrasonic volume data by CHI or THI using an ultrasonic diagnostic apparatus in order to reproduce the preliminary examination and determine the final approach position of the ultrasonic probe 1. Even at the time of reproduction of such preliminary examination, it is necessary to align the ultrasonic volume data at the time of preliminary examination and the ultrasonic volume data at the time of reproduction of preliminary examination (use case 2).

  Whether or not the ablation area (predicted ablation area) predicted from the arrival position of the puncture needle sufficiently includes the boundary area between the liver cancer and the liver cancer when the puncture needle reaches the planned position even when RFA is performed. For example, ultrasonic volume data is generated by THI. Even at the time of evaluation of such an expected ablation area, for example, alignment of X-ray CT volume data at the time of surgical procedure determination or ultrasonic volume data at the time of reproduction of preliminary examination and ultrasonic volume data at the time of evaluation of the expected ablation area (Use case 3).

  Even after the RFA is performed, ultrasonic volume data is generated in order to evaluate the RFA ablation area. Further, for example, after the completion of the ablation, generation of ultrasonic volume data by CHI or contrast X-ray CT is performed. Volume data is generated. Even at the time of evaluating the RFA ablation area, the ultrasonic volume data generated immediately after the RFA is performed, the contrast ultrasound volume data generated 5 minutes after the end of the ablation, and the contrast X-ray CT volume data are aligned. Needed (use case 4).

  Even during the follow-up examination, the doctor performs, for example, a contrast X-ray CT examination or a contrast ultrasound examination in order to determine the therapeutic effect. Even during the follow-up examination, it is necessary to align the volume data at the follow-up examination with, for example, the ultrasonic volume data at the time of evaluating the expected ablation area (use case 5). Further, in the case of use case 5, all the volume data used in use cases 1 to 4 can be subject to alignment.

  In the use case described above, the operator designates two volume data for performing comparative interpretation. Specifically, the operator designates the latest ultrasonic volume data generated by each use case and the volume data to be compared. In other words, the operator designates two volume data, at least one of which is ultrasonic volume data. In the following, a case where two pieces of ultrasonic volume data generated by the ultrasonic diagnostic apparatus according to the present embodiment are designated as an alignment target will be described as an example.

  For example, in use case 3, the operator designates ultrasonic volume data at the time of reproducing the preliminary examination and ultrasonic volume data at the time of evaluating the expected ablation area. In the present embodiment, the ultrasonic volume data generated by the ultrasonic diagnostic apparatus shown in FIG. 1 and the volume data generated by another medical image diagnostic apparatus (for example, X-ray CT volume data at the time of determining a technique). ) Is specified even if it is specified. In such a case, the X-ray CT volume data at the time of determining the technique is transferred from the external device 4 to the internal storage unit 16.

  Then, the target section receiving unit 17a, the calculation unit 17b, the extraction unit 17c, the installation unit 17d, and the alignment unit 17e included in the volume data processing unit 17 illustrated in FIG. 1 perform the following processing.

  The target cross-section receiving unit 17a inputs two cross-sections (target cross-sections) that include a common structure in two volume data, which is ultrasonic volume data generated based on reflected ultrasonic waves, to the input device 3. It accepts from an operator via.

  The calculating unit 17b divides the first cross section, which is one of the two target cross sections received by the target cross section receiving unit 17a, into a plurality of sub-fractions, and calculates the feature amounts of the plurality of sub-fractions. To do.

  The extraction unit 17c extracts a region of interest (ROI) for aligning two volume data from the plurality of small fractions based on the feature amounts of the plurality of small fractions calculated by the calculation unit 17b.

  The alignment unit 17e performs alignment of two volume data using the ROI extracted by the extraction unit 17c. Specifically, the alignment unit 17e performs alignment using the ROI pair installed by the installation unit 17d. That is, the installation unit 17d has the same positional relationship in the first cross section of the ROI extracted by the extraction unit 17c as the second cross section that is the other target cross section received by the target cross section receiving unit 17a together with the first cross section. Set up a paired ROI. Then, the alignment unit 17e aligns the two volume data using the ROI pair set as a pair for each of the two volume data by the processing of the extraction unit 17c and the installation unit 17d.

  More specifically, the alignment unit 17e maintains a relative position with respect to a corresponding target cross section of each ROI set as a pair for each of two volume data by the processing of the extraction unit 17c and the installation unit 17d. Any one of the first cross section and the second cross section is shifted to a plurality of positions. Then, each time the two alignment sections are moved, the alignment unit 17e calculates the similarity between the regions of interest using an evaluation function, and based on the positional relationship between the regions of interest in which the calculation result is optimized. The two volume data are aligned.

  A specific example of the processing of the volume data processing unit 17 described above will be described below with reference to FIGS. 4 and 5 are diagrams for explaining the volume data processing unit shown in FIG.

  First, under the control of the control unit 18, ultrasonic volume data (reference volume data shown in FIG. 4) at the time of reproduction of the preliminary examination and ultrasonic volume data (target volume data shown in FIG. 4) at the time of evaluation of the expected ablation area. Are displayed on the monitor 2. Specifically, first, the image generation unit 14 generates, for example, an MPR image at an initial position in the orthogonal three-axis coordinate system set in the volume data under the control of the control unit 18. Then, the monitor 2 displays the MPR image at the initial position of the reference volume data and the MPR image at the initial position of the target volume data, as shown in FIG.

  Then, the operator searches for a structure that is common to the reference volume data and the target volume data, that is, a main target structure that is a main target in positioning. Specifically, the operator operates the “translation movement button” and the “rotation movement button” displayed together with the MPR image at the initial position on the monitor 2 using, for example, a mouse included in the input device 3. Then, the position of the cross section for generating the MPR image is changed. As a result, the operator selects and displays the target cross section including the main target structure as shown in FIG. That is, the operator selects and displays the MPR image in which the main target structure is depicted by operating the “translation movement button” and the “rotation movement button”.

  Then, for example, the operator designates the main target structure drawn on each MPR image selected and displayed as shown in FIG. 4C using the mouse of the input device 3. Thereby, the target cross section of the reference volume data and the target cross section of the target volume data are determined. As a result, the target cross section receiving unit 17a receives the target cross section of the reference volume data and the target cross section of the target volume data. Specifically, the target cross section receiving unit 17a receives the position of the target cross section in the reference volume data and the position of the target cross section in the target volume data.

  Then, the calculation unit 17b divides the first cross section, which is one of the two target cross sections received by the target cross section receiving unit 17a, into a plurality of subfractions, and the feature amounts of the plurality of subfractions, respectively. Is calculated. For example, the calculation unit 17b sets the target cross section of the reference volume data as the first cross section, and divides the first cross section into a plurality of subfractions so that the subfractions are overlapped. And the calculation part 17b calculates "average information amount (entropy): H" as a feature-value by the following formula | equation (1), for example. Note that “pi” shown in Expression (1) is a relative frequency at which the pixel value “i” in the small fraction appears in the small fraction.

  In Equation (1), the base of the logarithm is “2”, but an arbitrary numerical value such as “10” can be set as the base of the logarithm when calculating the average information amount.

  Then, the extraction unit 17c extracts the ROI based on the feature amount (average information amount) of each of the plurality of small fractions calculated by the calculation unit 17b. For example, the extraction unit 17c arranges the average information amounts of the plurality of small fractions calculated by the calculation unit 17b in descending order, and extracts the top three small fractions from the first to the third as ROIs. In the example shown in FIG. 3D, the result of extracting three ROIs by the extraction unit 17c in the target cross section of the reference volume data is shown.

  In the present embodiment, a small fraction with the maximum average information amount is extracted as an ROI, or a small fraction with an average information amount greater than a predetermined value is extracted as an ROI. Even if it exists, it is applicable.

  Then, as shown in (D) of FIG. 3, the installation unit 17d has the same cross section in the first cross section of the three ROIs extracted by the extraction unit 17c as the target cross section of the target volume data. Three regions of interest that are paired with the cross section are installed. Thereby, the ROI pair is set for each of the two volume data by the processing of the extraction unit 17c and the installation unit 17d. That is, the ROI pair is set for each of the two volume data in the main target structure or in the vicinity of the main target structure by the processing of the extraction unit 17c and the installation unit 17d.

  This embodiment is applicable even when the target cross section of the target volume data is set as the first cross section and the target cross section of the reference volume data is set as the second cross section. The feature amount calculated by the calculation unit 17b is not limited to the average information amount, and any index value can be used as long as a characteristic structure depicted in the target cross section can be extracted. it can.

  Then, the alignment unit 17e calculates the similarity between the ROI pairs using an evaluation function that calculates the mutual information amount. Further, the alignment unit 17e shifts the position of the second cross section to a plurality of positions, for example, in a state where the relative position of each ROI with respect to the corresponding target cross section is maintained. The alignment unit 17e calculates the mutual information amount as the similarity between the ROIs each time the two target cross sections are moved. For example, as shown in (A) to (D) of FIG. 5, the alignment unit 17 e has a desired movement amount set in advance such as “initial setting, shift 1, shift 2, shift 3”. The second cross section is translated or rotated.

Then, the alignment unit 17e aligns the two volume data based on the positional relationship between the ROIs in which the mutual information value has been optimized. That is, the alignment unit 17e determines the amount of positional deviation between the reference volume data and the target volume data from the coordinate parameter of the second cross section in which the mutual information amount value is optimized.
Thereby, the monitor 2 displays the best alignment result as shown in FIG. Specifically, the monitor 2 displays two MPR images in which substantially the same cross section of the subject P is depicted.

Here, processing executed by the alignment unit 17e will be described below using mathematical formulas and the like. The ROI in the first cross section is “R”, and the ROI in the second cross section is “F”. In addition, a two-dimensional histogram generated from the pixel values (r, f) of the pixels at the corresponding positions between “R” and “F” is defined as “H _RF (r, f)”. Further, the number of pixels of “R” and “F” is “V”. At this time, the probability density function “P _RF (r, f)” taken by the pixel values “R” and “F” is obtained by the following equation (2).

Further, “P _R ” which is a probability density function “P _R (r)” taken by the pixel value “ _R ” is obtained by the following equation (3).

Further, “P _F ” which is a probability density function “P _F (f)” taken by the pixel value “ _F ” is obtained by the following equation (4).

  The mutual information “T (F; R)” is obtained by the following equation (5) from the probability density function obtained by the equations (2) to (4).

  That is, the alignment unit 17e calculates the mutual information amount between the ROIs of the first cross section and the second cross section using these equations (1) to (5). The alignment unit 17e moves the position of the ROI in the second cross section to a plurality of locations by a predetermined amount at each shift position, and calculates the mutual information amount at each location. Then, the alignment unit 17e sets the maximum value of the calculated mutual information amount as “the mutual information amount of the shift position”.

  The alignment unit 17e calculates the mutual information amount each time the second cross section is moved. Here, for coordinate conversion accompanying translation or rotational movement of the second cross section, at least six coordinate parameters of translational movement and rotational movement in the X direction, Y direction, and Z direction, and if necessary, three shear directions are also included. Nine coordinate parameters are used. When this coordinate conversion is “α”, the ROI of the second cross section after the coordinate conversion is expressed as “α (F)”. Therefore, the alignment unit 17e determines α that maximizes the mutual information “T (α (F); R)” between “R” and “α (F)” using a known optimization method. Thus, the amount of positional deviation between the reference volume data and the target volume data is determined.

  For example, the alignment unit 17e determines the amount of displacement between the reference volume data and the target volume data by performing the shifting process of the second cross section until the mutual information amount becomes the optimum value standard. Alternatively, the alignment unit 17e performs the second cross-section shifting process a predetermined number of times (for example, seven times), and based on the position of the second cross-section where the mutual information amount is maximized, the reference volume data and the target volume data The amount of misalignment between and may be determined. Note that the target cross section to be shifted may be the first cross section.

  As described above, the alignment unit 17e is configured so that the ultrasonic volume data generated between the two ultrasonic volume data generated by the ultrasonic diagnostic apparatus according to the present embodiment, the ultrasonic volume data generated by the ultrasonic diagnostic apparatus according to the present embodiment, and the like. Registration with the volume data generated by the medical image diagnostic apparatus. Therefore, the alignment unit 17e can perform alignment between a plurality of volume data via one ultrasonic volume data. As a result, the monitor 2 can display a plurality of aligned volume data. FIG. 6 is a diagram for explaining an example of the processing result of the alignment unit.

  For example, as a result of processing by the alignment unit 17e, as shown in FIG. 6, the monitor 2 uses ultrasonic volume data generated by “pre-treatment THI” and ultrasonic volume data generated by “pre-treatment CHI”. And MPR images at substantially the same position of the subject P with ultrasonic volume data generated by “in-treatment THI” and X-ray CT volume data generated by “post-treatment contrast X-ray CT examination” can do. The operator moves the cross-sectional position of the MPR image in an aligned state by operating the “translation movement button” and the “rotation movement button” displayed together with each MPR image on the display screen. Can do.

  In the present embodiment, other evaluation functions may be used in addition to the function for calculating the mutual information amount as the evaluation function for the alignment unit 17e to calculate the similarity. For example, the alignment unit 17e uses, as an evaluation function for calculating the similarity, a cross-correlation coefficient, a root mean square of luminance difference values between ROIs, or a difference value of luminance (pixel values) between ROIs. A function for calculating the average amount of information may be used. Even using these similarities, the alignment unit 17e can perform alignment between two volume data by moving the first cross section or the second cross section.

  Next, processing of the ultrasonic diagnostic apparatus according to the present embodiment will be described with reference to FIGS. FIG. 7 is a flowchart for explaining the ROI setting process of the ultrasonic diagnostic apparatus according to this embodiment. FIG. 8 is a flowchart for explaining the alignment process of the ultrasonic diagnostic apparatus according to this embodiment.

  As shown in FIG. 7, the ultrasonic diagnostic apparatus according to the present embodiment has received designation of reference volume data and target volume data, at least one of which is ultrasonic volume data, from the operator via the input device 3. Is determined (step S101). Here, when the designation of the reference volume data and the target volume data is not accepted (No at Step S101), the ultrasonic diagnostic apparatus enters a standby state.

  On the other hand, when the designation of the reference volume data and the target volume data is received (Yes at Step S101), the control unit 18 reads the designated reference volume data and the target volume data, and sends them to the image generation unit 14 and the monitor 2. Control is performed to generate and display an MPR image at the initial position from each of the two volume data (step S102).

  Then, the target cross-section receiving unit 17a determines whether or not the designation of the main target structure has been received in each MPR image displayed on the monitor 2 (step S103). Here, when the designation of the main target structure is not accepted (No at Step S103), the control unit 18 determines whether or not an initial position change request is accepted (Step S104). Here, when the request for changing the initial position is not received (No at Step S104), the target cross-section receiving unit 17a performs the determination process at Step S103 according to an instruction from the control unit 18.

  On the other hand, when the request for changing the initial position is received (Yes at step S104), the control unit 18 sends the MPR image at the position (cross section) changed from each of the two volume data to the image generation unit 14 and the monitor 2. Control to generate and display (step S105). And according to the instruction | indication of the control part 18, the object cross-section reception part 17a performs the determination process of step S103.

  On the other hand, when the designation of the main target structure is received (Yes at Step S103), the target cross section receiving unit 17a performs the target cross section (first cross section) in the reference volume data in which the main target structure is specified and the target in the target volume data. The position information of the cross section (second cross section) is notified to the calculation unit 17b.

  Then, the calculation unit 17b divides the first cross-section among the two target cross sections received by the target cross-section receiving unit 17a (step S106), and calculates the average information amount of each divided sub-fraction. Calculate (step S107).

  Thereafter, the extraction unit 17c extracts an ROI (region of interest for aligning two volume data) in the first section based on the average information amount of each of the plurality of small fractions calculated by the calculation unit 17b. (Step S108).

  Subsequently, the installation unit 17d installs a pair of ROIs on the second cross section with the same positional relationship in the first cross section of the ROI extracted by the extraction unit 17c (step S109), and the setting of the ROI pair is completed. To do.

  Then, the alignment unit 17e aligns the two volume data using the ROI pair set as a pair for each of the two volume data by the processing of the extraction unit 17c and the installation unit 17d.

  That is, as shown in FIG. 8, the alignment unit 17e determines whether or not an ROI pair has been set for each of the two volume data by the processing of the extraction unit 17c and the installation unit 17d (step S201). Here, when the ROI pair is not set (No at Step S201), the alignment unit 17e stands by until the ROI pair is set.

  On the other hand, when the ROI pair is set (Yes at Step S201), the alignment unit 17e calculates the mutual information amount between the ROIs (Step S202), and determines whether or not the calculated mutual information amount satisfies the optimum value criterion. Determination is made (step S203). If the optimum value criterion is not satisfied (No at Step S203), the alignment unit 17e moves the second cross section, that is, performs coordinate conversion of the second cross section (Step S204), and moves the ROI on the second cross section. (Step S205). Then, the alignment unit 17e returns to step S202, and calculates the mutual information amount between the ROIs using the moved RIO.

  On the other hand, when the optimum value criterion is satisfied (Yes at Step S203), the alignment unit 17e determines the amount of positional deviation between the reference volume data and the target volume data (Step S206). Thereby, the alignment between the reference volume data and the target volume data is completed.

  Then, the control unit 18 performs control so that the aligned volume data is displayed in parallel on the monitor 2 (see step S207, (E) of FIG. 5), and ends the process.

  As described above, in the present embodiment, the target cross-section receiving unit 17a includes a common structure in the two volume data, at least one of which is ultrasonic volume data generated based on the reflected wave of the ultrasonic wave. Two cross sections (target cross sections) are received from the operator via the input device 3. The calculating unit 17b divides the first cross section, which is one of the two target cross sections received by the target cross section receiving unit 17a, into a plurality of sub-fractions, and calculates the feature amounts of the plurality of sub-fractions. To do. The extraction unit 17c extracts a region of interest (ROI) for aligning two volume data from the plurality of small fractions based on the feature amounts of the plurality of small fractions calculated by the calculation unit 17b. The alignment unit 17e performs alignment of two volume data using the ROI extracted by the extraction unit 17c.

  That is, in this embodiment, the ROI can be set based on the feature amount in the cross section including the main target structure that is determined to be important for the operator to perform alignment. In other words, in this embodiment, it is possible to set an ROI that can be more accurately aligned than the main target structure designated by the operator. Therefore, in this embodiment, it is possible to perform accurate alignment between volume data. In the present embodiment, the operator can change the slice position of the MPR image displayed on the monitor 2 and simply perform the selection and designation of the main target structure to perform the alignment automatically. Therefore, it is possible to reduce the burden on the operator related to the alignment process.

  Moreover, in this embodiment, the installation part 17d is the other target cross section which the target cross section reception part 17a received with the said 1st cross section in the same positional relationship in the 1st cross section of ROI extracted by the extraction part 17c. A pair of regions of interest are set on a second cross section. Then, the alignment unit 17e aligns the two volume data using the ROI pair set as a pair for each of the two volume data by the processing of the extraction unit 17c and the installation unit 17d.

  Therefore, in this embodiment, since the ROI pair is automatically set, it is possible to further reduce the burden on the operator related to the alignment process.

  Further, in the present embodiment, the alignment unit 17e maintains the relative position with respect to the corresponding target cross section of each ROI set as a pair for each of the two volume data by the processing of the extraction unit 17c and the installation unit 17d. Thus, either the first cross section or the second cross section is shifted to a plurality of positions. Then, each time the two alignment sections are moved, the alignment unit 17e calculates the similarity between the regions of interest using an evaluation function, and based on the positional relationship between the regions of interest in which the calculation result is optimized. The two volume data are aligned.

  That is, in this embodiment, since the alignment is performed while shifting one position of the ROI pair, it is possible to avoid the alignment by the local solution. As a result, in the present embodiment, it is possible to improve the alignment accuracy between the volume data.

  In the present embodiment, the calculation unit 17b calculates an average information amount as a feature amount, and the alignment unit 17e calculates a mutual information amount as a similarity. That is, in the present embodiment, alignment can be performed using a known technique, and accurate alignment between volume data can be easily performed.

  Note that the above-described embodiment may be executed by a modification described below. Hereinafter, a modification according to the present embodiment will be described with reference to FIG. FIG. 9 is a diagram for explaining a modification according to the present embodiment.

  In the embodiment described above, a case where the first cross-section is divided into two-dimensional sub-fractions out of two target cross-sections including the main target structure, and the feature amount of each two-dimensional sub-fraction is calculated. explained. However, in the present embodiment, as shown in FIG. 9A, the calculation unit 17b separates the three-dimensional first region including the first cross section into a three-dimensional sub-fraction, It may be a case where a fraction is calculated. In such a case, the extraction unit 17c extracts a three-dimensional ROI from the first region. Then, the installation unit 17d installs a pair of ROIs in the second region including the second cross section in the same positional relationship as the positional relationship in the first region of the three-dimensional ROI extracted by the extraction unit 17c. Then, the alignment unit 17e calculates a similarity (for example, mutual information amount) using a three-dimensional ROI pair, and aligns the reference volume data and the target volume data.

  By applying a modification example in which a three-dimensional ROI pair is installed, the amount of information used for alignment can be increased compared to a two-dimensional ROI pair, and the alignment between volume data can be more accurately performed. Can be performed.

  In the above-described embodiment, a case has been described in which a plurality of ROI pairs set in two volume data by the processing of the extraction unit 17c and the installation unit 17d are used as they are for the alignment processing. However, the present embodiment may be a case where the alignment unit 17e aligns two volume data using the ROI pair designated by the operator. For example, as shown in FIG. 9B, the operator clicks the ROI (a) extracted in the first section displayed on the monitor 2 using the mouse of the input device 3. Then, the ROI in the second cross section corresponding to ROI (a) and ROI (a) is excluded from the ROI for alignment. In such a case, the alignment unit 17e aligns the two volume data using two ROI pairs other than those excluded by the operator.

  For example, the doctor who performs comparative interpretation interprets and excludes the ROI pair set at a position away from the main structural object because it is not necessary to perform accurate alignment.

  Alternatively, when performing alignment processing between volume data shot by different FOV (Field Of View), the operator designates an ROI pair used for alignment. For example, the ROI installed in the second cross section of the ultrasonic volume data as a ROI that forms a pair with the ROI extracted in the first cross section of the X-ray CT volume data is outside the FOV of the ultrasonic volume data. In such a case, even if such an ROI pair is used, the alignment accuracy is lowered. Therefore, the operator excludes the ROI pair set outside the FOV of one volume data, and designates the ROI pair set within the FOV of both volume data as the ROI pair used for alignment.

  As described above, according to the modification in which the designation of the ROI pair is left to the operator, the volume data is aligned according to the intention of the operator, and the accuracy of the alignment between the volume data is prevented from being lowered. It becomes possible.

  In the above-described embodiment, the case where the alignment process is performed in the ultrasonic diagnostic apparatus has been described. However, the above-described alignment process may be performed by an image processing apparatus installed independently of the ultrasonic diagnostic apparatus. Specifically, the image processing apparatus having the display control functions of the volume data processing unit 17 and the control unit 18 shown in FIG. 1 is a medical image diagnostic apparatus including an ultrasonic diagnostic apparatus, a PACS database, an electronic medical record, or the like. There may be a case where two volume data received from the database of the system are received and the above-described alignment process is performed.

  The image processing method described in the above-described embodiment can be realized by executing an image processing program prepared in advance on an image processing apparatus such as a personal computer or a workstation. This image processing program can be distributed via a network such as the Internet. The image processing program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, and a DVD, and is read from the recording medium by an image processing apparatus that is a computer. It can also be executed.

  As described above, according to the present embodiment, accurate positioning between volume data can be performed.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 Ultrasonic probe 2 Monitor 3 Input device 4 External apparatus 10 Apparatus main body 11 Transmission / reception part 12 B mode process part 13 Doppler process part 14 Image generation part 15 Image memory 16 Internal storage part 17 Volume data process part 17a Target cross-section reception part 17b Calculation Unit 17c extraction unit 17d installation unit 17e positioning unit 18 control unit 19 interface unit 100 network

Claims (9)

A receiving unit that receives two target cross sections including a common structure from an operator via a predetermined input unit in two volume data, at least one of which is volume data generated based on an ultrasonic reflected wave;
A first cross section that is one of the two target cross sections received by the receiving unit, or a first region that includes the first cross section is divided into a plurality of subfractions, and the plurality of subfractions A calculation unit for calculating each feature amount;
An extraction unit that extracts a region of interest for aligning the two volume data from the plurality of small fractions based on the feature amounts of the plurality of small fractions calculated by the calculation unit;
An alignment unit that aligns the two volume data using the region of interest extracted by the extraction unit;
An ultrasonic diagnostic apparatus comprising: The region of interest paired with the second cross section, which is the other target cross section received by the receiving unit together with the first cross section, in the same positional relationship in the first cross section of the region of interest extracted by the extracting unit. Installation that installs a region of interest that is paired with the second region including the second cross section in the same positional relationship as the positional relationship in the first region of the region of interest extracted by the extraction unit Further comprising
The alignment unit performs alignment of the two volume data using a region-of-interest pair set as a pair for each of the two volume data by the processing of the extraction unit and the installation unit. The ultrasonic diagnostic apparatus according to claim 1.   When a plurality of region-of-interest pairs are set as a pair for each of the two volume data by the processing of the extraction unit and the installation unit, the alignment unit uses the region-of-interest pair designated by the operator, The ultrasonic diagnostic apparatus according to claim 2, wherein two volume data are aligned.   The alignment unit is configured to maintain the relative position of each region of interest set as a pair for each of the two volume data by the processing of the extraction unit and the installation unit with respect to a corresponding target cross section. After shifting either the cross-section or the second cross-section to a plurality of positions, the similarity between the regions of interest is calculated using a predetermined evaluation function, and the regions of interest between which the calculation results are optimized are calculated. The ultrasonic diagnostic apparatus according to claim 2 or 3, wherein the two volume data are aligned based on a positional relationship.   The alignment unit, as the evaluation function for calculating the similarity, a mutual information amount, a cross-correlation coefficient, a root mean square of luminance difference values between regions of interest, or between regions of interest The ultrasonic diagnostic apparatus according to claim 4, wherein a function that calculates an average information amount of luminance difference values is used.   The ultrasonic diagnostic apparatus according to claim 1, wherein the calculation unit calculates an average information amount as the feature amount.   The ultrasonic diagnostic apparatus according to claim 1, wherein the calculation unit divides the first region into three-dimensional small fractions. A receiving unit that receives two target cross sections including a common structure from an operator via a predetermined input unit in two volume data, at least one of which is volume data generated based on an ultrasonic reflected wave;
A first cross section that is one of the two target cross sections received by the receiving unit, or a first region that includes the first cross section is divided into a plurality of subfractions, and the plurality of subfractions A calculation unit for calculating each feature amount;
An extraction unit that extracts a region of interest for aligning the two volume data from the plurality of small fractions based on the feature amounts of the plurality of small fractions calculated by the calculation unit;
An alignment unit that aligns the two volume data using the region of interest extracted by the extraction unit;
An image processing apparatus comprising: An acceptance procedure for receiving two target cross sections including a common structure from an operator via a predetermined input procedure in two volume data, at least one of which is volume data generated based on an ultrasonic reflected wave,
The first cross-section that is one of the two target cross-sections received by the receiving procedure or the first region including the first cross-section is divided into a plurality of sub-fractions, and the plurality of sub-fractions A calculation procedure for calculating each feature amount;
An extraction procedure for extracting a region of interest for aligning the two volume data from the plurality of small fractions based on the feature quantities of the plurality of small fractions calculated by the calculation procedure;
An alignment procedure for aligning the two volume data using the region of interest extracted by the extraction procedure;
An image processing program for causing a computer to execute.

Images